[evlatests] timing tic / Freaky Fridays.

[Vivek Dhawan]

 From the previous message: "The main problem, the occasional time offset
on the whole array, is still with us. It was last seen on Feb 21; the one
before on Feb 14."

On Friday eve, Feb 28 UT, baselines and pointing were run in anticipation
of trouble. Sure enough, both those datasets show a timing error.

The current theory is that the weekly EOP updates are implicated.

The details may not matter now but I may as well set them down:

PROJECT  DATE     start UT stop  comment

AE172 2009-02-27  19:40 - 04:38  Q band, shows clearly that the trouble
                                  starts at 0h UT.

BASEL 2009-02-28  04:46 - 05:38  delta T=0.72s.

POINT 2009-02-28  05:54 - 07:39  delta T=0.85s.

AM975 2009-02-28  07:40 - 09:38  K,Q bands, del T ~0. Clean again!

Vivek.